On Fri, Oct 16, 2015 at 11:44:17AM +0200, Thomas Schwinge wrote:
> But: working on getting our changes into trunk, for example, when we make
> an effort to extract from gomp-4_0-branch self-contained, individual
> patches, but it then takes weeks to get commit approval or review
> comments, I don't see how that's going to work for the thousands of lines
> patches that we're still to submit? I mean, GCC development stage 1 is
> going to end in just a few weeks, I suppose?
Yes, see Richi's mail from today, there are 3 weeks plus weekend left in
stage 1. I'm sorry for the delays in patch review, and will try to arrange
an hour or two daily for OpenACC/OpenMP/HSA patch review for the remainder of
stage1 (and still work on the missing OpenMP 4.5 features or fixes during
the rest of the days). The PTX changes Nathan can review himself, or Bernd.
And I certainly appreciate if Bernd reviews some of the OMP bits from time
to time, after all, all global reviewers can do that.
> also, allow for "somewhat non-perfect" changes to be committed, and later
> address the "warts"? (Allowing for incremental progress, while keeping
I'd prefer not to do that though, that will just mean the bad code will stay
around forever.
In any case, if you have unreviewed OpenACC/OpenMP/libgomp patches ready to
be merged into trunk, please ping them.
Jakub
